ABSTRACT ? PROJECT 3: UW-CNOF BIOLOGICAL VALIDATION DEVELOPMENT The complexity of hierarchical interactions within the nucleus demands that easy-to-use and accurate methods for mapping and modeling both close-range and long-range interactions be developed. In this project, the experimental and computational methods developed in Projects 1 and 2 will be put to the test to evaluate their ability to detect interactions at high resolution, as well as for the prediction of the sequence determinants of specific aspects of genome architecture. In Aim 1, we will apply newly developed methods to well-defined mouse and human biological systems in which the 4D structure of specific genomic regions and chromosomes can be anticipated. Our validation strategy will employ systems in which alleles can be identified to facilitate studies of diploid cells, i.e. tissues and cell lines from a mouse interspecific cross. Interactions between loci will be tested at enhancer/promoter regions, while interactions at topologically associated domains (TADs) will be tested by comparing the active and inactive X chromosomes in female cells. This functional validation approach will be complemented by high resolution DNA-FISH analyses to verify specific interactions. In Aim 2, to validate our approaches for generating a 4D view of dynamic changes in nuclear structure, we will measure interactions in single cells during the cell cycle and during mouse myoblast and embryonic stem (ES) cell differentiation. By focusing on relatively well understood aspects of these systems, we will achieve validation by linking dynamic changes in the nucleome to other layers of regulation. Analysis of mouse ESC differentiation will exploit the wealth of knowledge that surrounds X inactivation, a process central to nuclear remodeling in mammals, while skeletal myoblast differentiation is well characterized with respect to its transcriptional regulation. In Aim 3, we will validate our ability to predict the sequence determinants of genome architecture. Specifically, we will perform controlled manipulations of defined genomic regions, either by allele- specific heterozygous CRISPR/Cas9 targeting to generate cells with engineered deletions at promoter/enhancer interacting regions and at regions between TADs, or by Xist-mediated silencing of full autosomes. The biological validation work performed in all aims of this project will facilitate the progressive optimization of both bulk and single cell DNase Hi-C protocols (Project 1) and new approaches to modeling the 4D nucleome (Project 2), while also paving the way for biological model development and data generation (Project 4).